The study analyzed the electronic health records (EHR) of all surgical patients at the University of Miami Hospital from October 2014 through April 2015. The first four months of data collection (N=1,794) was abstracted from the EHR to test the measurement and structural model using structural equation modeling (SEM) to develop a prediction tool utilizing the regression coefficients. Validation utilized an independent sample to confirm the prediction tools ability to predict severe postoperative pain. Results supported the independent predictors of baseline heart rate, preoperative pain intensity, expected surgical pain (low, moderate, high, highest), age, female gender, and Hispanic ethnicity were significant factors in predicting the presence of severe postoperative pain on postoperative day 1.
